Introduction and business problem¶

Brooklyn is one of the most diverse places in New York. Here live over 30% of NY population. However, when it comes to vegetarian food, not many places offer good deals. Vegetarianism is more welcome among young population. In this project I’ll come up with some possible location options for the venue.

The problem to solve:

Open a vegetarian restaurant of the mid-range price category in Brooklyn. In specific, I decided to take a baseline location point the NYC College of Technology. The place is close to subway stations (as will be presented below), and has not got many dining venues so far. Therefore, the stakeholders are mostly young people preferring vegetarian food and local employees in vicinity.

Data¶

In the framework of this project, I’ll be using several data sources, including:

New York neighborhood data as per NY neighborhood data
NYC real estate sales data (https://geo.nyu.edu/catalog/nyu-2451-34678)
Foursquare api service (https://api.foursquare.com/)
NY subway stations data (https://geo.nyu.edu/catalog/nyu-2451-34503)
NY universities locations (https://geo.nyu.edu/catalog/nyu-2451-34491)

These will be used to build the neighborhood profile in vicinity and better identify if the area worth starting a business at. For instance, NYC real estate data will give the first insights on the price range to buy a place if needed, as well as will give some insights on the solvency of the local population. While foursquare data will help build the profile of the restaurants in vicinity and identify some potential competitors. University locations will help determine if the starting point location (to be discussed in the main report methodology section) presents the point of university agglomeration. Subway stations will be used to illustarte good commute options.

From the sources above, with exception of foursqaure, was downloaded as shapefiles and, since not everywhere JSON option was working, lateron converted to csv using mygeodata service: https://mygeodata.cloud/converter/shp-to-csv

Methodology¶

The purpose of this project is to identify the possible locations for a vegetarian restaurant in Brooklyn, NYC. The project presents an active use of spatial data from NYU Spatial data repository.

The initial prespotion for research is mid-range real estate pricings bring mid-range restaurants category. This assumption was tested in the course of research. At the first stage, the geodata for NY real estate will be taken and segmented (using k-means clustering) to map the distribution of building types sold in 2016. The data is to some extent outdated, yet for the purposes of this project, given no major price shifts in the previous years, the data will show the structural distribution of the sales objects.

Once done, the clusters will be mapped using the folio library.

Then, we'll use the foursquare data to get the list and details of some venues in vicinity of 2 km. This will allow us to have a more comoprehensive look and build the places profile based on their price tier, location, tips, likes, rating and other similar charachteristics. Once done, the second clastrization will be made, this time of the venues in question. This step will help better locate the competitors. The data will be illustarted visually respecitvely.

On the third stage I'll examine if any other educational instituions are in vicinity and how good the commute is. Once that all is put together, some street names for possible restaurant openning will be presented.

The starting point for the location analysis is NYC College of Technology. This place is taken with the view to reaching yough people and employees who work nearby, thus covering the target audience.

#import necessary libraries 
import pandas as pd
pd.set_option('display.max_columns', None)
pd.set_option('display.max_rows', None)

import numpy as np

import json # library to handle JSON files

#!conda install -c conda-forge geopy --yes # uncomment this line if you haven't completed the Foursquare API lab
from geopy.geocoders import Nominatim # convert an address into latitude and longitude values

import requests # library to handle requests
from pandas.io.json import json_normalize # tranform JSON file into a pandas dataframe

# Matplotlib and associated plotting modules
import matplotlib.cm as cm
import matplotlib.colors as colors

# import k-means from clustering stage
from sklearn.cluster import KMeans

#!conda install -c conda-forge folium=0.5.0 --yes # uncomment this line if you haven't completed the Foursquare API lab
import folium # map rendering library

print('Libraries imported.')

Libraries imported.

with open('ny_data/newyork_data.json') as json_data:
    newyork_data = json.load(json_data)

newyork_data['features'][0]

{'type': 'Feature',
 'id': 'nyu_2451_34572.1',
 'geometry': {'type': 'Point',
  'coordinates': [-73.84720052054902, 40.89470517661]},
 'geometry_name': 'geom',
 'properties': {'name': 'Wakefield',
  'stacked': 1,
  'annoline1': 'Wakefield',
  'annoline2': None,
  'annoline3': None,
  'annoangle': 0.0,
  'borough': 'Bronx',
  'bbox': [-73.84720052054902,
   40.89470517661,
   -73.84720052054902,
   40.89470517661]}}

# define the dataframe columns
column_names = ['Borough', 'Neighborhood', 'Latitude', 'Longitude'] 

# instantiate the dataframe
neighborhoods = pd.DataFrame(columns=column_names)

for data in newyork_data['features']:
    borough = neighborhood_name = data['properties']['borough'] 
    neighborhood_name = data['properties']['name']
        
    neighborhood_latlon = data['geometry']['coordinates']
    neighborhood_lat = neighborhood_latlon[1]
    neighborhood_lon = neighborhood_latlon[0]
    
    neighborhoods = neighborhoods.append({'Borough': borough,
                                          'Neighborhood': neighborhood_name,
                                          'Latitude': neighborhood_lat,
                                          'Longitude': neighborhood_lon}, ignore_index=True)

print('The dataframe has {} boroughs and {} neighborhoods.'.format(
        len(neighborhoods['Borough'].unique()),
        neighborhoods.shape[0]) )

The dataframe has 5 boroughs and 306 neighborhoods.

neighborhoods['Borough'].unique()

array(['Bronx', 'Manhattan', 'Brooklyn', 'Queens', 'Staten Island'],
      dtype=object)

brooklyn_data = neighborhoods[neighborhoods['Borough'] == 'Brooklyn'].reset_index(drop=True)
brooklyn_data.head()

#open the real estate data file

with open('ny_data/nyu-2451-34678-geojson.json') as json_data:
    newyork_sale_ft_data = json.load(json_data)

sales_data = json_normalize(newyork_sale_ft_data['features'])

sales_data.columns[0].split('.')[1]

'coordinates'

columns = sales_data.columns
cols = []
for i in columns:
    if len(i.split('.')) > 1:
        cols.append(i.split('.')[1])
    else:
        cols.append(i.split('.')[0])
        
sales_data.columns = cols
cols

['coordinates',
 'type',
 'geometry_name',
 'id',
 'address',
 'apt',
 'bbl_id',
 'bbox',
 'bldg_cls_p',
 'bldg_cls_s',
 'bldg_ctgy',
 'block',
 'borough',
 'com_unit',
 'easmnt',
 'georesult',
 'land_sqft',
 'lat',
 'long',
 'lot',
 'message',
 'nbhd',
 'price',
 'res_unit',
 'sale_date',
 'sale_id',
 'tax_cls_p',
 'tax_cls_s',
 'tot_sqft',
 'tot_unit',
 'usable',
 'year',
 'yr_built',
 'zip',
 'type']

sales_data.sample()

Clustering the neighbourhoods on real estate sales¶

Let's first clasterize neighbourhoods of Brooklyn by real estate sales data to get some first insights on the neighbourhood economic activity

In data, borough 3 is Brooklyn, and data will be taken only where sale price > 10 USD, i.e. usuable buildings.

brooklyn_property_sales = sales_data[(sales_data['borough'] == 3) & (sales_data['usable'] == 'Y')] 

brooklyn_property_sales = brooklyn_property_sales[['borough','nbhd', 'address', 'zip', 'lat', 'long', 'bldg_ctgy', 'bldg_cls_s', 'tax_cls_s', 'land_sqft','price']]

brooklyn_property_sales.bldg_cls_s.value_counts()

# make onehot encodding for selected columns

brooklyn_property_sales = pd.get_dummies(brooklyn_property_sales, prefix='CAT_', columns= ['bldg_ctgy', 'bldg_cls_s',] )


brooklyn_property_sales.head()

address = 'Brooklyn, NY'

geolocator = Nominatim(user_agent="ny_explorer")
location = geolocator.geocode(address)
latitude = location.latitude
longitude = location.longitude
print('The geograpical coordinate Brooklyn borough are {}, {}.'.format(latitude, longitude))

The geograpical coordinate Brooklyn borough are 40.6501038, -73.9495823.

# create map of Brooklyn using latitude and longitude values
map_brooklyn = folium.Map(location=[latitude, longitude], zoom_start=11)

# add markers to map
for lat, lng, label in zip(brooklyn_property_sales['lat'], brooklyn_property_sales['long'], brooklyn_property_sales['nbhd']):
    label = folium.Popup(label, parse_html=True)
    folium.CircleMarker(
        [lat, lng],
        radius=5,
        popup=label,
        color='blue',
        fill=True,
        fill_color='#3186cc',
        fill_opacity=0.7,
        parse_html=False).add_to(map_brooklyn)  


#add New York City College of Technology to the map (big dark circle)
label = folium.Popup('New York City College of Technology', parse_html=True)
folium.CircleMarker(
        [40.695457, -73.9864678851903],
        radius=20,
        popup=label,
        color='dark green',
        fill=True,
        fill_color='dark green',
        fill_opacity=0.7,
        parse_html=False).add_to(map_brooklyn)     
    
map_brooklyn

brooklyn_property_sales_cluster = brooklyn_property_sales.iloc[:,6:]
brooklyn_property_sales_cluster.head()

from sklearn.cluster import KMeans
from sklearn import metrics
from scipy.spatial.distance import cdist
import numpy as np
import matplotlib.pyplot as plt
from sklearn.preprocessing import StandardScaler

# scale the data and make clusterization
se = StandardScaler()


X = se.fit_transform(brooklyn_property_sales_cluster)

# k means determine k
distortions = []
K = range(1,10)
for k in K:
    kmeanModel = KMeans(n_clusters=k).fit(X)
    kmeanModel.fit(X)
    distortions.append(sum(np.min(cdist(X, kmeanModel.cluster_centers_, 'euclidean'), axis=1)) / X.shape[0])

# Plot the elbow
plt.plot(K, distortions, 'bx-')
plt.xlabel('k')
plt.ylabel('Distortion')
plt.title('The Elbow Method showing the optimal k')
plt.show()

/usr/local/anaconda3/lib/python3.7/site-packages/sklearn/preprocessing/data.py:645: DataConversionWarning: Data with input dtype uint8, int64, object were all converted to float64 by StandardScaler.
  return self.partial_fit(X, y)
/usr/local/anaconda3/lib/python3.7/site-packages/sklearn/base.py:464: DataConversionWarning: Data with input dtype uint8, int64, object were all converted to float64 by StandardScaler.
  return self.fit(X, **fit_params).transform(X)

<Figure size 640x480 with 1 Axes>

# let's stick with 5 clusters

kclusters = 5

# run k-means clustering
kmeans = KMeans(n_clusters=kclusters, random_state=0).fit(X)

# check cluster labels generated for each row in the dataframe
kmeans.labels_[0:10]

array([1, 1, 1, 1, 1, 1, 1, 1, 1, 1], dtype=int32)

len(kmeans.labels_)

14797

brooklyn_property_sales['clusters'] = kmeans.labels_

brooklyn_property_sales['clusters'] = brooklyn_property_sales['clusters'].dropna().astype(int)

# create map
map_clusters = folium.Map(location=[latitude, longitude], zoom_start=11)

# set color scheme for the clusters
x = np.arange(kclusters)
ys = [i + x + (i*x)**2 for i in range(kclusters)]
colors_array = cm.rainbow(np.linspace(0, 1, len(ys)))
rainbow = [colors.rgb2hex(i) for i in colors_array]

# add markers to the map
markers_colors = []
for lat, lon, poi, cluster in zip(brooklyn_property_sales['lat'], brooklyn_property_sales['long'], brooklyn_property_sales['nbhd'], brooklyn_property_sales['clusters']):
    label = folium.Popup(str(poi) + ' Cluster ' + str(cluster), parse_html=True)
    folium.CircleMarker(
        [lat, lon],
        radius=5,
        popup=label,
        color=rainbow[cluster-1],
        fill=True,
        fill_color=rainbow[cluster-1],
        fill_opacity=0.7).add_to(map_clusters)

    
#add New York City College of Technology to the map (big dark circle)  
label = folium.Popup('New York City College of Technology', parse_html=True)
folium.CircleMarker(
        [40.695457, -73.9864678851903],
        radius=20,
        popup=label,
        color='dark green',
        fill=True,
        fill_color='dark green',
        fill_opacity=0.7,
        parse_html=False).add_to(map_clusters)    
       
map_clusters

Analyzing the clusters¶

brooklyn_property_sales['clusters'].value_counts()

0    7297
4    5346
1    2083
2      62
3       9
Name: clusters, dtype: int64

Cluster 0¶

brooklyn_property_sales[brooklyn_property_sales['clusters'] == 0][:10]

Cluster 1¶

brooklyn_property_sales[brooklyn_property_sales['clusters'] == 1][:10]

Cluster 2¶

brooklyn_property_sales[brooklyn_property_sales['clusters'] == 2][:10]

Cluster 3¶

brooklyn_property_sales[brooklyn_property_sales['clusters'] == 3][:10]

Cluster 4¶

brooklyn_property_sales[brooklyn_property_sales['clusters'] == 4][:10]

Having analyzed the clusters by value counts of the category values, here's my interptretation of each of them:

cluster 0, COOPS - ELEVATOR APARTMENTS, CONDOS - ELEVATOR APARTMENTS and RENTALS - WALKUP APARTMENTS with price between 550,000 USD and 1,200,000 USD. The most populous cluster

cluster 1, ONE FAMILY DWELLINGS, with price varying between 450,000 USD and 800,000 USD on average

cluster 2, RENTALS, with price varying between 150,000 USD and 500,000 USD on average

cluster 3, CONDOS, with varying price, just several instances

cluster 4, ONE FAMILY DWELLINGS to THREE FAMILY DWELLINGS, with price between roughly 650,000 USD and 1,200,000 USD

Also, let's get the cluster distribution from the brooklyn neighborhoods and pivot it to get the list of unique neighborhoods. Then, we'll merge these data with brooklyn_data file containing the coordinates and neighborhood names.

brooklyn_property_pivot = brooklyn_property_sales.pivot_table(index = 'nbhd', columns = 'clusters', values = 'zip', aggfunc = 'count')
brooklyn_property_pivot.index = brooklyn_property_pivot.index.str.title()
brooklyn_property_pivot.head()

# merge with brooklyn_data. Since the number of nbhds is slighly lower on the propertt sales data, the inner join will cut some of them in the resultsing df.

brooklyn_data_merged = brooklyn_data.merge(brooklyn_property_pivot, left_on = 'Neighborhood', right_on = 'nbhd')
cols = ['Borough', 'Neighborhood', 'Latitude', 'Longitude', 'sc_0', 'sc_1', 'sc_2', 'sc_3', 'sc_4']
brooklyn_data_merged.columns = cols
brooklyn_data_merged.head()

Having clustered Brooklyn on real estate data and cluster distribution by neighborhood, we see that our starting point, New York City College of Technology, is located in in cluster 0.
This corresponds to mid to high range price of real estate sales. Just what's needed!

Foursquare database¶

CLIENT_ID = '-' # your Foursquare ID
CLIENT_SECRET = '-' # your Foursquare Secret
VERSION = '20180605' # Foursquare API version

print('Your credentails:')
print('CLIENT_ID: ' + CLIENT_ID)
print('CLIENT_SECRET:' + CLIENT_SECRET)

Your credentails:
CLIENT_ID: -
CLIENT_SECRET:-

def getNearbyVenues(latitudes, longitudes, radius=2000, LIMIT = 300):
    
    venues_list=[]    
       
    # create the API request URL - restaurants
    category = '4d4b7105d754a06374d81259'


    url = 'https://api.foursquare.com/v2/venues/explore?client_id={}&client_secret={}&v={}&ll={},{}&categoryId={}&radius={}&limit={}'.format(
        CLIENT_ID, 
        CLIENT_SECRET, 
        VERSION, 
        neighborhood_latitude, 
        neighborhood_longitude,
        category,
    
        radius, 
        LIMIT)
      
      # make the GET request
    results = requests.get(url).json()["response"]['groups'][0]['items']
      
      # return only relevant information for each nearby venue
    venues_list.append([(
            v['venue']['id'], 
            v['venue']['name'],
            v['venue']['categories'][0]['name'],
            v['venue']['location']['lat'], 
            v['venue']['location']['lng'],
            v['venue']['location']['distance']          
            ) for v in results])

    nearby_venues = pd.DataFrame([item for venue_list in venues_list for item in venue_list])
    nearby_venues.columns = [ 
                  'ID',
                  'Venue', 
                  'Venue Category',
                  'Venue Latitude', 
                  'Venue Longitude',
                  'Distance from the NYC College of Technology',
                  ]
    
    return(nearby_venues)

# coordinates of the NYC College of Technology 

neighborhood_latitude = 40.695457
neighborhood_longitude =  -73.9864678851903

nb = getNearbyVenues(neighborhood_latitude, neighborhood_longitude)
nb.set_index('ID')

nb.shape

(100, 6)

nb.groupby('Venue Category')['Venue'].count().sort_values(ascending = False)

Venue Category
Bakery                           8
American Restaurant              7
Pizza Place                      7
Italian Restaurant               7
Bagel Shop                       6
Japanese Restaurant              6
French Restaurant                5
Thai Restaurant                  3
Burger Joint                     3
Middle Eastern Restaurant        3
Sandwich Place                   3
Deli / Bodega                    3
New American Restaurant          2
Mediterranean Restaurant         2
Café                             2
Seafood Restaurant               2
Sushi Restaurant                 2
Chinese Restaurant               2
Asian Restaurant                 2
Restaurant                       2
Diner                            1
Burrito Place                    1
Cajun / Creole Restaurant        1
Bistro                           1
Caribbean Restaurant             1
Creperie                         1
Cuban Restaurant                 1
Vegetarian / Vegan Restaurant    1
Dumpling Restaurant              1
Falafel Restaurant               1
Food Court                       1
Food Truck                       1
Gastropub                        1
Israeli Restaurant               1
Kebab Restaurant                 1
Latin American Restaurant        1
Mexican Restaurant               1
Peruvian Restaurant              1
Salad Place                      1
South American Restaurant        1
Spanish Restaurant               1
Taco Place                       1
German Restaurant                1
Name: Venue, dtype: int64

Now, let's get more details about these places, including the rating and the price tier

details = []

for i in nb.ID:
    url = 'https://api.foursquare.com/v2/venues/{}?client_id={}&client_secret={}&v={}&'.format(
        i,
        CLIENT_ID, 
        CLIENT_SECRET, 
        VERSION 
        )
    results = requests.get(url).json()
    
    try:
        details.append([
                    results['response']['venue']['id'] if results['response']['venue']['id'] else 0,
                    results['response']['venue']['stats']['tipCount'] if results['response']['venue']['stats']['tipCount'] else 0,
                    results['response']['venue']['price']['tier'] if results['response']['venue']['price'] else 0,
                    results['response']['venue']['likes']['count'] if results['response']['venue']['likes']['count'] else 0,
                    results['response']['venue']['rating'] if results['response']['venue']['rating'] else 0 ])
    except KeyError:
        continue

nearby_venues = pd.DataFrame(details)
nearby_venues.columns = [ 
                  'ID',
                  'Tips_count', 
                  'Price tier', 
                  'Likes_count',
                  'Rating',
                  ]
nearby_venues.set_index('ID')

Price tier¶

From the foursquare description, currently the valid range of price points are [1,2,3,4], 1 being the least expensive, 4 being the most expensive. For food venues, in the United States, 1 is < 10 USD an entree, 2 is 10-20 USD an entree, 3 is 20-30USD an entree, 4 is > 30 USD an entree

nb_merged = nb.merge(nearby_venues, right_index = True, left_index = True).drop('ID_y',axis = 1)
nb_merged

Let's add these places to the brooklyn map and clusterize them thereafter.

address = 'Brooklyn, NY'

geolocator = Nominatim(user_agent="ny_explorer")
location = geolocator.geocode(address)
latitude = location.latitude
longitude = location.longitude
print('The geograpical coordinate of Brooklyn are {}, {}.'.format(latitude, longitude))

The geograpical coordinate of Brooklyn are 40.6501038, -73.9495823.

# create map of Brooklyn using latitude and longitude values
map_brooklyn_1 = folium.Map(location=[latitude, longitude], zoom_start=11)

# add markers to map
for lat, lng, label in zip(nb['Venue Latitude'], nb['Venue Longitude'], nb['Venue']):
    label = folium.Popup(label, parse_html=True)
    folium.CircleMarker(
        [lat, lng],
        radius=5,
        popup=label,
        color='blue',
        fill=True,
        fill_color='#3186cc',
        fill_opacity=0.7,
        parse_html=False).add_to(map_brooklyn_1)  


#add New York City College of Technology to the map
label = folium.Popup('New York City College of Technology', parse_html=True)
folium.CircleMarker(
        [40.695457, -73.9864678851903],
        radius=20,
        popup=label,
        color='dark green',
        fill=True,
        fill_color='dark green',
        fill_opacity=0.7,
        parse_html=False).add_to(map_brooklyn_1)     
    
map_brooklyn_1

Okay, that's a good sign since most of the venues are on the opposite side of the river. Let's clusterize them.

venues_cluster = nb_merged.iloc[:,5:]


se = StandardScaler()


X_1 = se.fit_transform(venues_cluster)

# k means determine k
distortions = []
K = range(1,10)
for k in K:
    kmeanModel = KMeans(n_clusters=k).fit(X_1)
    kmeanModel.fit(X_1)
    distortions.append(sum(np.min(cdist(X_1, kmeanModel.cluster_centers_, 'euclidean'), axis=1)) / X_1.shape[0])

# Plot the elbow
plt.plot(K, distortions, 'bx-')
plt.xlabel('k')
plt.ylabel('Distortion')
plt.title('The Elbow Method showing the optimal k')
plt.show()

/usr/local/anaconda3/lib/python3.7/site-packages/sklearn/preprocessing/data.py:645: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.
  return self.partial_fit(X, y)
/usr/local/anaconda3/lib/python3.7/site-packages/sklearn/base.py:464: DataConversionWarning: Data with input dtype int64, float64 were all converted to float64 by StandardScaler.
  return self.fit(X, **fit_params).transform(X)

# let's stick with 3 clusters

kclusters = 3

# run k-means clustering
kmeans = KMeans(n_clusters=kclusters, random_state=0).fit(X_1)

# check cluster labels generated for each row in the dataframe
kmeans.labels_[0:10]

array([0, 2, 2, 2, 2, 2, 2, 2, 2, 2], dtype=int32)

pd.Series(kmeans.labels_).value_counts()

2    42
1    29
0    26
dtype: int64

nb_merged['cluster'] = kmeans.labels_
nb_merged.head()

Add clesters labels to the map

# create map
map_clusters_1 = folium.Map(location=[latitude, longitude], zoom_start=11)

# set color scheme for the clusters
x = np.arange(kclusters)
ys = [i + x + (i*x)**2 for i in range(kclusters)]
colors_array = cm.rainbow(np.linspace(0, 1, len(ys)))
rainbow = [colors.rgb2hex(i) for i in colors_array]

# add markers to the map
markers_colors = []
for lat, lon, poi, cat, cluster in zip(nb_merged['Venue Latitude'], nb_merged['Venue Longitude'], nb_merged['Venue'], nb_merged['Venue Category'] ,  nb_merged['cluster']):
    label = folium.Popup(str(poi) + "_" + str(cat) + ' Cluster ' + str(cluster), parse_html=True)
    folium.CircleMarker(
        [lat, lon],
        radius=5,
        popup=label,
        color=rainbow[cluster-1],
        fill=True,
        fill_color=rainbow[cluster-1],
        fill_opacity=0.7).add_to(map_clusters_1)

#add New York City College of Technology to the map   
label = folium.Popup('New York City College of Technology', parse_html=True)
folium.CircleMarker(
        [40.695457, -73.9864678851903],
        radius=20,
        popup=label,
        color='dark green',
        fill=True,
        fill_color='dark green',
        fill_opacity=0.7,
        parse_html=False).add_to(map_clusters_1)    
map_clusters_1

Quick analysis of the venues clusters¶

Cluster 0¶

nb_merged[nb_merged['cluster']  == 0].groupby('Venue Category')['Venue'].count().sort_values(ascending = False)

Venue Category
Pizza Place                      3
Italian Restaurant               2
Bagel Shop                       2
Deli / Bodega                    2
American Restaurant              2
Sandwich Place                   2
New American Restaurant          1
Mexican Restaurant               1
Kebab Restaurant                 1
Japanese Restaurant              1
Seafood Restaurant               1
French Restaurant                1
Food Court                       1
Restaurant                       1
Cajun / Creole Restaurant        1
Burrito Place                    1
Bakery                           1
Asian Restaurant                 1
Vegetarian / Vegan Restaurant    1
Name: Venue, dtype: int64

From here, we exclude several categories: Pizza place, Bagel Shop, Kebab, Seafood, Bakery, Sandwich since they're not our competitors.

nb_merged.info()

<class 'pandas.core.frame.DataFrame'>
Int64Index: 97 entries, 0 to 96
Data columns (total 11 columns):
ID_x                                           97 non-null object
Venue                                          97 non-null object
Venue Category                                 97 non-null object
Venue Latitude                                 97 non-null float64
Venue Longitude                                97 non-null float64
Distance from the NYC College of Technology    97 non-null int64
Tips_count                                     97 non-null int64
Price tier                                     97 non-null int64
Likes_count                                    97 non-null int64
Rating                                         97 non-null float64
cluster                                        97 non-null int32
dtypes: float64(3), int32(1), int64(4), object(3)
memory usage: 8.7+ KB

nb_merged[(nb_merged['cluster']  == 0) & (~(nb_merged['Venue Category'].isin(['Pizza Place','Bagel Shop', 
                                                                                 'Kebab Restaurant', 
                                                                                 'Seafood Restaurant',
                                                                                 'Sandwich Place'])))]

Cluster 1¶

nb_merged[nb_merged['cluster']  == 1].groupby('Venue Category')['Venue'].count().sort_values(ascending = False)

Venue Category
American Restaurant          4
Italian Restaurant           3
Bakery                       2
Japanese Restaurant          2
French Restaurant            2
Gastropub                    1
Burger Joint                 1
Café                         1
Deli / Bodega                1
Dumpling Restaurant          1
Food Truck                   1
Thai Restaurant              1
German Restaurant            1
Taco Place                   1
Mediterranean Restaurant     1
Middle Eastern Restaurant    1
Pizza Place                  1
Salad Place                  1
Spanish Restaurant           1
Sushi Restaurant             1
Israeli Restaurant           1
Name: Venue, dtype: int64

From here, we exclude several categories: Bakery, Pizza place, Burger Joint, Sushi Restaurant, since they're not our competitors.

nb_merged[(nb_merged['cluster']  == 1) & (~(nb_merged['Venue Category'].isin(['Bakery', 'Pizza Place', 'Burger Joint', 'Sushi Restaurant'])))]

Cluster 2¶

nb_merged[nb_merged['cluster']  == 2].groupby('Venue Category')['Venue'].count().sort_values(ascending = False)

Venue Category
Bakery                       5
Japanese Restaurant          3
Bagel Shop                   3
Thai Restaurant              2
Burger Joint                 2
Chinese Restaurant           2
Italian Restaurant           2
French Restaurant            2
Pizza Place                  2
New American Restaurant      1
Restaurant                   1
Asian Restaurant             1
South American Restaurant    1
Seafood Restaurant           1
Bistro                       1
Sandwich Place               1
Café                         1
Caribbean Restaurant         1
Creperie                     1
Middle Eastern Restaurant    1
Cuban Restaurant             1
Diner                        1
Falafel Restaurant           1
Sushi Restaurant             1
Peruvian Restaurant          1
Latin American Restaurant    1
Mediterranean Restaurant     1
American Restaurant          1
Name: Venue, dtype: int64

From here, we exclude several categories: Bakery, Bagel Shop, Pizza place, Burger Joint, Seafood Restaurant , Sushi Restaurant, Restaurant since they're not our competitors.

nb_merged[(nb_merged['cluster']  == 1) & (~(nb_merged['Venue Category'].isin(['Bakery', 'Pizza Place', 
                                                                              'Burger Joint', 'Sushi Restaurant',
                                                                             'Bagel Shop', 'Seafood Restaurant', 'Restaurant' ])))]

From the analysis above we can conclude that:

cluster 0 - cheap and mid range venues with low to moderate tips count and high middle rating, close vicinity (till 800-900 m), mostly Bakery and Asian venues
cluster 1 - cheap to mid range venues with moderate tips count, moderate likes count, and middle rating, middle distance (800-1300 m) mostly American and European restaurants
cluster 2 - expensive venues with middle to huge tips count, likes count, and high rating, middle distance(700-1000 m), mostly American and European restaurants

Since our restaurant is going to be vegetarian and middle range income, our direct competitors would be mostly cluster 0 venues and possibly cluster 1, even if they're further located. There's even one direct competitor, another vegie venue - Wild Ginger, in cluster 2, located about 900 m from the college. But it's a more expensive venue, which can be of advantage to us.

Let's see our top 20 competitors:

comp = nb_merged[((nb_merged['cluster']  == 0) | (nb_merged['cluster']  == 1)) 
          & (~(nb_merged['Venue Category'].isin(['Bakery', 'Pizza Place', 
                                                 'Burger Joint', 'Sushi Restaurant',
                                                 'Bagel Shop', 'Seafood Restaurant', 
                                                 'Restaurant','Kebab Restaurant', 'Sandwich Place' ])))][['Venue', 'Venue Category',
                                        'Distance from the NYC College of Technology',
                                        'Tips_count',
                                        'Price tier', 
                                        'Rating' ] ].nlargest(20, columns = 'Rating').reset_index(drop = True)

comp

Metro stations¶

Let's look for the nearby metro stations around the college

ny_metro_stations = pd.read_csv('ny_data/ny_metro_stations.csv')[['stop_name','complex_id','stop_lat','stop_lon' ]].dropna()

brooklyn_metro_stations = ny_metro_stations[ny_metro_stations['complex_id'].str.startswith('bk')]
brooklyn_metro_stations.head()

# add markers to map
for lat, lng, label in zip(brooklyn_metro_stations['stop_lat'], brooklyn_metro_stations['stop_lon'], brooklyn_metro_stations['stop_name']):
    label = folium.Popup(label, parse_html=True)
    folium.CircleMarker(
        [lat, lng],
        radius=10,
        popup=label,
        color='grey',
        fill=True,
        fill_color='grey',
        fill_opacity=0.7,
        parse_html=False).add_to(map_clusters_1)    
        
map_clusters_1

Excellent! In the vicinity we have three metro stations: Court St., Borough hall, and Jay St MetroTech.

University locations¶

Now, let's get the data of the New York universities to identify if the chosen starting point is actually part of the educational conglomerate

ny_uni = pd.read_csv('ny_data/nyu_uni.csv')
ny_uni.head()

def get_location(address):
    address = address + ', New York'
    try:
        geolocator = Nominatim(user_agent="ny_explorer")
        location = geolocator.geocode(address)
        latitude = location.latitude
        longitude = location.longitude
        results = pd.DataFrame([address,latitude,longitude])
        return latitude, longitude
    except:
        pass

get_uni_location = ny_uni.address.apply(get_location)

ny_uni['latitude'] = [i[0] if i is not None else None for i in get_uni_location]
ny_uni['longitude'] = [i[1] if i is not None else None for i in get_uni_location]

ny_uni = ny_uni.dropna()[['name', 'latitude','longitude']]

# add universities to map
for lat, lng, label in zip(ny_uni['latitude'], ny_uni['longitude'], ny_uni['name']):
    label = folium.Popup(label, parse_html=True)
    folium.CircleMarker(
        [lat, lng],
        radius=10,
        popup=label,
        color='yellow',
        fill=True,
        fill_color='yellow',
        fill_opacity=0.7,
        parse_html=False).add_to(map_clusters_1) 
    
map_clusters_1

Results and Discussion¶

The purpose of this project has been to investigate and segment the neighbourhoods in Brooklyn. In specific, with respect to opening the vegetarian restaurant in vicinity of NYC College of Technology. The starting point has been taken with respect to proximity to the metro stations, high number of young people who are more likely to consume organic / vegetarian food. The requirement was to open the spot in middle range income area. For this purpose I analyzed a more macro indicator of real estate sales, revealing that our terriroty indeed is a mid income range one. The analysis presented that most of the nearby buildings where flats were sold, belong to the cluster 0, which corresponds to the COOPS - ELEVATOR APARTMENTS, CONDOS - ELEVATOR APARTMENTS and RENTALS - WALKUP APARTMENTS with price between 550,000 USD and 1,200,000 USD. This is the most populous cluster.

Second, i segmented the restaurant venues in proximity of 2 km, in order to reach the optimal distribution / number of places. It turned out that most of the direct competitors would be American and European restaurants plus Pizza places, mid range restaurants. Since our restaurant is going to be vegetarian and middle range income, our direct competitors would be mostly cluster 0 and 1 venues, even if they're further located. There's even one direct competitor, another vegie venue - Wild Ginger, in cluster 2, located about 900 m from the college. But it's a more expensive venue, which can be of advantage to us. The College of Technology indeed appeared to be a part of educational conglomerate as shown on the map, with 5 instituions presented in total, namely: NYC College of Technology, St Francis College,Brooklyn Law School, ASA College, and Institute of Design and Construction.

Therefore, I recommend to locate a new restaurant nearby Jay Street, Adams Street, Fulton Mall, Boroghby street, and Tillary Street. Also, an alternative could be on Remsen street, near the St Francis College. It's close to the subway station at Court St., and has no many competitors too close.

Conclusion¶

In the end, the possible locations for a new vegetarian restaurant in Brooklyn are nearby Jay Street, Adams Street, Fulton Mall, Boroghby street, and Tillary Street. These streets present a good commute options, not far from the universities. Some other smaller streets in vicinity such as Chapel st. could be taken as well, however we must make sure it's not directly near the cathedral like there.

The project has been quite a challenge, however i am glad i finished it. Hope that my findings at least present a healthy way of analysis and are a good starting point for a more indepth analysis to the potential investor who indeed would open a restaurant like this in Brooklyn.

The project necesaarily had certain limitations which are related with the lack of the current data, for example, on the real estate sales. And in a commercial project some more thourough investigation would be necessary.

	Borough	Neighborhood	Latitude	Longitude
0	Brooklyn	Bay Ridge	40.625801	-74.030621
1	Brooklyn	Bensonhurst	40.611009	-73.995180
2	Brooklyn	Sunset Park	40.645103	-74.010316
3	Brooklyn	Greenpoint	40.730201	-73.954241
4	Brooklyn	Gravesend	40.595260	-73.973471

	borough	nbhd	address	zip	lat	long	tax_cls_s	land_sqft	price	CAT__01 ONE FAMILY DWELLINGS	CAT__A1	CAT__A5	CAT__S1
6291	3	BATH BEACH	8647 15TH AVENUE	11228	40.610414	-74.010528	1	1547	758000	1	0	1	0
6292	3	BATH BEACH	55 BAY 10TH STREET	11228	40.609857	-74.009897	1	1933	778000	1	0	1	0
6294	3	BATH BEACH	1906 86TH STREET	11214	40.605798	-74.000248	1	1900	1365000	1	0	0	1
6295	3	BATH BEACH	50 BAY 23RD STREET	11214	40.604094	-74.000011	1	2417	750000	1	1	0	0
6296	3	BATH BEACH	1964 86TH STREET	11214	40.604966	-73.998862	1	1725	1470000	1	0	0	1

	tax_cls_s	land_sqft	price	CAT__01 ONE FAMILY DWELLINGS	CAT__A1	CAT__A5	CAT__S1
6291	1	1547	758000	1	0	1	0
6292	1	1933	778000	1	0	1	0
6294	1	1900	1365000	1	0	0	1
6295	1	2417	750000	1	1	0	0
6296	1	1725	1470000	1	0	0	1

	borough	nbhd	address	zip	lat	long	tax_cls_s	land_sqft	price	CAT__07 RENTALS - WALKUP APARTMENTS	CAT__09 COOPS - WALKUP APARTMENTS	CAT__C1	CAT__C2	CAT__C3	CAT__C6	CAT__C7
6499	3	BATH BEACH	8674 17TH AVENUE	11214	40.607297	-74.006220	2	2432	1250000	1	0	0	1	0	0	0
6504	3	BATH BEACH	15 BAY 23RD STREET	11214	40.604558	-73.999499	2	2417	1330000	1	0	0	0	1	0	0
6505	3	BATH BEACH	8672 21ST AVENUE	11214	40.602010	-73.997389	2	2433	1380000	1	0	0	0	1	0	0
6507	3	BATH BEACH	45 BAY 28TH STREET	11214	40.601928	-73.996237	2	5800	5040000	1	0	1	0	0	0	0
6509	3	BATH BEACH	8717 21ST AVENUE	11214	40.600874	-73.998542	2	6767	3195000	1	0	0	0	0	0	1
6510	3	BATH BEACH	8758 BAY PARKWAY	11214	40.598953	-73.996936	2	9691	15150000	1	0	0	0	0	0	1
6514	3	BATH BEACH	1859-1861 CROPSEY AVENUE	11214	40.601738	-74.005708	2	3814	1000000	1	0	0	0	0	0	1
6515	3	BATH BEACH	194 BAY 22ND STREET	11214	40.601645	-74.003720	2	5051	3100000	1	0	1	0	0	0	0
6516	3	BATH BEACH	141 BAY 20TH, 3	11214	40.603632	-74.004044	2	0	350000	0	1	0	0	0	1	0
6517	3	BATH BEACH	139 BAY 20TH, 1	11214	40.603663	-74.004015	2	0	420000	0	1	0	0	0	1	0

	borough	nbhd	address	zip	lat	long	tax_cls_s	land_sqft	price	CAT__01 ONE FAMILY DWELLINGS	CAT__A1	CAT__A2	CAT__A5	CAT__S1	clusters
6291	3	BATH BEACH	8647 15TH AVENUE	11228	40.610414	-74.010528	1	1547	758000	1	0	0	1	0	1
6292	3	BATH BEACH	55 BAY 10TH STREET	11228	40.609857	-74.009897	1	1933	778000	1	0	0	1	0	1
6294	3	BATH BEACH	1906 86TH STREET	11214	40.605798	-74.000248	1	1900	1365000	1	0	0	0	1	1
6295	3	BATH BEACH	50 BAY 23RD STREET	11214	40.604094	-74.000011	1	2417	750000	1	1	0	0	0	1
6296	3	BATH BEACH	1964 86TH STREET	11214	40.604966	-73.998862	1	1725	1470000	1	0	0	0	1	1
6297	3	BATH BEACH	1970 86TH STREET	11214	40.604914	-73.998776	1	1725	1790000	1	0	0	0	1	1
6298	3	BATH BEACH	160 BAY 10TH STREET	11228	40.607955	-74.011906	1	2469	920000	1	0	0	1	0	1
6299	3	BATH BEACH	1638 BENSON AVENUE	11214	40.607659	-74.007902	1	3625	820000	1	1	0	0	0	1
6305	3	BATH BEACH	239 BAY8TH STREET	11228	40.607224	-74.015079	1	5810	1850000	1	0	1	0	0	1
6306	3	BATH BEACH	8804 17TH AVENUE	11214	40.604516	-74.009115	1	3394	1192000	1	0	0	0	1	1

	borough	nbhd	address	zip	lat	long	tax_cls_s	land_sqft	price	CAT__14 RENTALS - 4-10 UNIT	CAT__S4	clusters
6557	3	BATH BEACH	1934 BATH AVENUE	11214	40.602093	-74.002769	2	2160	1200000	1	1	2
7457	3	BAY RIDGE	8617 4 AVENUE	11209	40.622327	-74.028533	2	1720	108000	1	1	2
7458	3	BAY RIDGE	8615 4 AVENUE	11209	40.622373	-74.028515	2	1859	108000	1	1	2
8916	3	BEDFORD STUYVESANT	912 GATES AVENUE	11221	40.688822	-73.927879	2	1580	1100000	1	1	2
8925	3	BEDFORD STUYVESANT	529 MARCY AVENUE	11206	40.697131	-73.949548	2	2025	967698	1	1	2
8930	3	BEDFORD STUYVESANT	868 DEKALB AVENUE	11221	40.692518	-73.941322	2	2000	1450000	1	1	2
11041	3	BOROUGH PARK	4023 13 AVENUE	11218	40.639890	-73.987046	2	1615	990000	1	1	2
11045	3	BOROUGH PARK	5401 8 AVENUE	11220	40.638788	-74.006122	2	1613	4850000	1	1	2
11048	3	BOROUGH PARK	5917 10TH AVENUE	11219	40.633254	-74.004702	2	2000	1500000	1	1	2
11051	3	BOROUGH PARK	6005 FORT HAMILTON PARKWA	11219	40.632988	-74.005692	2	1495	2200000	1	1	2

	borough	nbhd	address	zip	lat	long	tax_cls_s	price	CAT__16 CONDOS - 2-10 UNIT WITH COMMERCIAL UNIT	CAT__R8	clusters
9045	3	BEDFORD STUYVESANT	186 FRANKLIN AVENUE	11205	40.693121	-73.957758	2	5000	1	1	3
9046	3	BEDFORD STUYVESANT	452 LAFAYETTE AVENUE	11205	40.689237	-73.956852	2	325840	1	1	3
10135	3	BOERUM HILL	438 ATLANTIC AVENUE	11217	40.686494	-73.983727	2	3250000	1	1	3
11169	3	BOROUGH PARK	1441 54TH STREET	11219	40.630323	-73.991858	2	566637	1	1	3
11172	3	BOROUGH PARK	857 60TH STREET	11220	40.634769	-74.008341	2	906620	1	1	3
12042	3	BROOKLYN HEIGHTS	253 HENRY STREET	11201	40.692750	-73.995262	2	500000	1	1	3
13134	3	BUSHWICK	245 HARMAN STREET	11237	40.698901	-73.919870	2	1260000	1	1	3
26055	3	PARK SLOPE SOUTH	443 7 AVENUE	11215	40.663130	-73.984706	2	299000	1	1	3
28361	3	WILLIAMSBURG-EAST	629 GRAND STREET	11211	40.711372	-73.946846	2	1313568	1	1	3

	borough	nbhd	address	zip	lat	long	tax_cls_s	land_sqft	price	CAT__02 TWO FAMILY DWELLINGS	CAT__B1	CAT__B2	CAT__B3	clusters
6322	3	BATH BEACH	8661 14TH AVENUE	11228	40.611605	-74.012876	1	2320	980000	1	1	0	0	4
6323	3	BATH BEACH	70 BAY 7TH STREET	11228	40.611048	-74.012271	1	2320	750000	1	1	0	0	4
6324	3	BATH BEACH	37 BAY 7 STREET	11228	40.611504	-74.011770	1	2320	970000	1	1	0	0	4
6325	3	BATH BEACH	35 BAY 7TH ST	11228	40.611531	-74.011741	1	2320	855000	1	1	0	0	4
6326	3	BATH BEACH	35 BAY 7TH STREET	11228	40.611531	-74.011741	1	2320	855000	1	1	0	0	4
6330	3	BATH BEACH	31 BAY 10TH STREET	11228	40.610305	-74.009429	1	2103	840000	1	0	0	1	4
6332	3	BATH BEACH	8653 16TH AVENUE	11214	40.609064	-74.008352	1	2058	769000	1	0	0	1	4
6335	3	BATH BEACH	18 BAY 14TH STREET	11214	40.608658	-74.006148	1	3033	1120075	1	1	0	0	4
6336	3	BATH BEACH	8659 17TH AVENUE	11214	40.607494	-74.005986	1	1933	375000	1	0	1	0	4
6337	3	BATH BEACH	8643 17TH AVENUE	11214	40.607725	-74.005744	1	1933	870000	1	0	0	1	4

clusters	0	1	2	3	4
nbhd
Bath Beach	55.0	20.0	1.0	NaN	110.0
Bay Ridge	375.0	134.0	2.0	NaN	163.0
Bedford Stuyvesant	431.0	44.0	3.0	2.0	527.0
Bensonhurst	57.0	42.0	NaN	NaN	158.0
Bergen Beach	16.0	42.0	NaN	NaN	66.0

	Venue	Venue Category	Venue Latitude	Venue Longitude	Distance from the NYC College of Technology
ID
594277b586f4cc0f251fc389	DeKalb Market Hall	Food Court	40.691250	-73.982579	571
59da9590e1f0aa52976b8f35	Han Dynasty	Chinese Restaurant	40.691334	-73.982456	570
5a87210eda5ede7ae86f82c6	Korilla BBQ	Asian Restaurant	40.693803	-73.985843	191
53d1a481498e1fac7c37f675	Pollo d'Oro	Peruvian Restaurant	40.694823	-73.983373	270
59a3040fda708024d6926771	Brooklyn Bridge Bistro And Winebar	Bistro	40.696272	-73.988343	182
4b6b6ad0f964a52066072ce3	Sushi Gallery	Sushi Restaurant	40.697595	-73.993236	618
53b36289498e528b6cad6624	Forno Rosso	Pizza Place	40.694437	-73.983442	279
5946d138e2da1964625ca9b2	Daigo Hand Roll Bar	Japanese Restaurant	40.691259	-73.982603	569
594416bca4ba7c683f5492df	A Taste Of Katz's	Sandwich Place	40.691378	-73.982426	567
50478b43e4b05a8c89d8198c	Sophies Cuban Cuisine	Cuban Restaurant	40.690602	-73.987700	550
479ccb47f964a5206b4d1fe3	Iron Chef House	Japanese Restaurant	40.697406	-73.992560	558
5949c93c3ba767778ec8fad9	Eight Turn Crêpe	Creperie	40.691186	-73.982831	565
52b204c0498efe3c3be32d91	Burrow	Bakery	40.702579	-73.986702	793
5c56083dc58ed7002c3255d0	Charm Kao	Thai Restaurant	40.689445	-73.986585	669
5c3ebf15c0f163002ce6db6f	CAVA	Mediterranean Restaurant	40.692461	-73.988529	376
4ac00013f964a520689320e3	Queen	Italian Restaurant	40.691319	-73.991647	635
55664aac498e10d4e5769e5e	Damascus Bread & Pastry Shop	Bakery	40.690047	-73.993054	819
583c7fa58cfe547612b47b93	Westville DUMBO	American Restaurant	40.702021	-73.989596	776
54b4026b498e73880a40a8d7	Grand Army	Seafood Restaurant	40.688329	-73.986612	793
4e793558aeb79f7dab5c7f6f	Sottocasa Pizzeria	Pizza Place	40.688534	-73.988947	798
5012043564a4944f5c47738a	Dellarocco's	Pizza Place	40.694992	-73.995924	799
4dc5a9c8e4cd169dc6532c66	Shelsky's of Brooklyn	Bagel Shop	40.689486	-73.992518	838
4dcc0c3052b17cba4fb62352	Five Guys	Burger Joint	40.693676	-73.986087	200
530931a5498e4079544a5f13	French Louie	French Restaurant	40.688056	-73.988159	836
3fd66200f964a520eae81ee3	Henry's End	New American Restaurant	40.699610	-73.991983	656
550c4a17498e1d6612e9dffc	Bread & Spread	Sandwich Place	40.702541	-73.987082	790
3fd66200f964a520ece81ee3	Noodle Pudding	Italian Restaurant	40.699715	-73.991769	651
43eb685af964a520382f1fe3	Yemen Cafe	Middle Eastern Restaurant	40.690026	-73.993579	851
51b0ceff2d4e2ee801fd4ab2	Foragers	Deli / Bodega	40.702468	-73.988569	800
5942d5c94c9be62191b036fa	sweetgreen	Salad Place	40.702914	-73.989830	877
49d3773df964a520fb5b1fe3	Vinegar Hill House	American Restaurant	40.702686	-73.981260	916
4a635ce2f964a520dbc41fe3	Lassen & Hennigs	Deli / Bodega	40.694970	-73.994857	710
59456df7a4ba7c683f19f0e7	Café D'Avignon	Café	40.691250	-73.982786	562
4b5e4c9df964a520648829e3	Mile End Delicatessen	Sandwich Place	40.687549	-73.987018	881
4a748cd9f964a52092de1fe3	Fast and Fresh Burrito Deli	Burrito Place	40.687983	-73.986935	832
4cbf097828d176b0a7c6226e	Colonie	American Restaurant	40.690733	-73.995963	958
49e63b62f964a52027641fe3	Ki Sushi	Sushi Restaurant	40.687574	-73.989936	925
49f9dd95f964a5208d6d1fe3	Wild Ginger	Vegetarian / Vegan Restaurant	40.687898	-73.989730	885
49f3261bf964a520696a1fe3	Clark's Restaurant	Diner	40.697533	-73.993044	601
4e1e23631838bed7eb6a8792	Bien Cuit	Bakery	40.687635	-73.989864	916
4f3265a119836c91c7d3c6f7	Hadramout Restaurant	Restaurant	40.689818	-73.993877	886
5c17e5658c35dc002cdc77e5	Chicks Isan	Thai Restaurant	40.690758	-73.983387	584
5a0740a3e1f22816d11723d5	Lillo	Italian Restaurant	40.690200	-73.996540	1032
585d72449f25836f2b2b7a1b	Xifu Food	Chinese Restaurant	40.688027	-73.982088	905
50ca6337e4b04e1f3135689c	Juliana's Pizza	Pizza Place	40.702769	-73.993616	1013
49e13617f964a520a9611fe3	One Girl Cookies	Bakery	40.687404	-73.990287	952
48a41073f964a52091511fe3	Hibino	Japanese Restaurant	40.690076	-73.996497	1037
5b7c4cb3c58ed7002c1fd9bd	Butler Bakeshop	Café	40.703295	-73.992526	1011
53cfc343498e6ad2a8cc30f3	Pio Bagel	Bagel Shop	40.692159	-73.986368	367
527d8cfc11d2a61a7c663024	Luzzo's BK	Pizza Place	40.690555	-73.995333	926
4be30f9dd27a20a1cd1f915b	River Deli	Italian Restaurant	40.693713	-73.998435	1028
43164480f964a52065271fe3	Bedouin Tent	Middle Eastern Restaurant	40.686936	-73.984469	963
4471bf9af964a5209c331fe3	Jack the Horse Tavern	American Restaurant	40.699940	-73.993639	784
5945bcb842d8c24270976841	Likkle More Jerk	Caribbean Restaurant	40.690743	-73.983388	585
4a9c07d1f964a520ca3520e3	Heights Falafel	Falafel Restaurant	40.698446	-73.992508	608
57d74b38498e41fcd6c3ee92	The Gumbo Bros	Cajun / Creole Restaurant	40.689526	-73.991730	795
593c0d2262420b7feccc3048	Cecconi's	Italian Restaurant	40.703893	-73.991609	1034
4f69f2b76d86f87117bb13ab	Gran Eléctrica	Mexican Restaurant	40.702570	-73.993096	969
4d9f5a9efc4f721e7e5a9d5f	Rucola	Italian Restaurant	40.685659	-73.985769	1092
538d22cb498e86974754d6da	Shake Shack	Burger Joint	40.703043	-73.994276	1071
5b2932a0f5e9d70039787cf2	Los Tacos Al Pastor	Taco Place	40.702436	-73.987539	782
580ba89038fa564f8dec12fb	Doner Kebab NYC	Kebab Restaurant	40.692116	-73.987493	381
5ca3681932b61d0039533330	Pret A Manger	French Restaurant	40.693555	-73.985378	230
5330e34e498e36f70456a24b	Saketumi Asian Bistro	Asian Restaurant	40.694910	-73.994578	687
4ab6d694f964a520407920e3	Henry Public	Gastropub	40.690413	-73.996412	1009
4a674aa8f964a5201fc91fe3	La Bagel Delight	Bagel Shop	40.691125	-73.991668	652
4f6e5ea9e4b086107787908b	La Vara	Spanish Restaurant	40.687851	-73.995582	1144
43504680f964a520b0281fe3	Almondine Bakery	Bakery	40.703328	-73.991253	964
5b7f53abe727c40024124d8c	Oh! Dumplings	Dumpling Restaurant	40.687722	-73.992943	1019
595183c718d43b1841112628	Dulcinea Churros & Co.	Bakery	40.691364	-73.982421	569
51a3ab412fc69e7654bc731f	Luke's Lobster	Seafood Restaurant	40.703441	-73.994091	1097
49e28bbcf964a5203a621fe3	Court Street Bagels	Bagel Shop	40.688026	-73.993060	996
4ae10c7af964a520db8421e3	My Little Pizzeria	Pizza Place	40.690236	-73.992334	763
3fd66200f964a520efe81ee3	The River Café	American Restaurant	40.703754	-73.994834	1162
59e7f722829b0c09cfce96d2	Black Forest Brooklyn	German Restaurant	40.685613	-73.991097	1163
3fd66200f964a52058f11ee3	Bar Tabac	French Restaurant	40.687369	-73.990106	951
51782c87498e1803b70a775f	DUMBO Food Truck Lot	Food Truck	40.703126	-73.986851	854
56a122d6498eb97a32f045d4	Maison Kayser	Bakery	40.692166	-73.991092	535
4db8989a8154ce84dc168e1c	two8two Bar & Burger	Burger Joint	40.688513	-73.989743	820
572be80d498e66d83ce6f537	Beasts & Bottles	New American Restaurant	40.690489	-73.995108	915
4f7f8b86e4b088077df30175	Chez Moi	French Restaurant	40.690654	-73.995777	950
58a202a25490d30f87553a08	Rice & Miso	Japanese Restaurant	40.684633	-73.983768	1226
4f21bb85e5e872143c0ca04b	One Girl Cookies	Bakery	40.703317	-73.990694	944
59399dcd12f0a93a5c392d6a	Kotti Berliner Döner Kebab	Restaurant	40.690662	-73.983533	588
4a5b37fff964a520edba1fe3	La Bagel Delight	Bagel Shop	40.702417	-73.988597	795
4a2aebabf964a5206e961fe3	Farmer in the Deli	Deli / Bodega	40.693276	-73.971837	1258
51dd7c69498ee00b70fa54b8	Atrium DUMBO	American Restaurant	40.703410	-73.990464	947
4112ca00f964a520ed0b1fe3	Joya	Thai Restaurant	40.686708	-73.993718	1150
59456723ad910e218eb07f49	Arepa Lady	South American Restaurant	40.691182	-73.982785	568
4ab11a2af964a5200b6820e3	Los Papis	Latin American Restaurant	40.702087	-73.985044	747
4bafd791f964a5203c253ce3	Mitoushi Sushi	Japanese Restaurant	40.690182	-73.993983	864
5250553911d262bb0c732ee8	Cafe Paulette	French Restaurant	40.689743	-73.975798	1102
5914d24b3b83070821fcdd87	Miss Ada	Israeli Restaurant	40.689560	-73.972351	1360
5739c66c498e7ef6085cec4f	Karasu	Japanese Restaurant	40.689577	-73.973290	1290
59dd280192e7a97276e4d52f	Celestine	Mediterranean Restaurant	40.704537	-73.988026	1019
5529a381498e1232f524541f	Numero 28 Pizza & Cucina	Italian Restaurant	40.686871	-73.990721	1020
49ba5f96f964a52060531fe3	Building on Bond	American Restaurant	40.686502	-73.985150	1003
4a9ff488f964a520b73d20e3	Tutt Cafe	Middle Eastern Restaurant	40.700285	-73.993422	795
5995a1ec08815835e6db45aa	Andrew’s Classic Brooklyn Bagels	Bagel Shop	40.690601	-73.983598	592
51fa84d22fc60e9c2e6f875b	Fornino	Pizza Place	40.692980	-74.001697	1314

	stop_name	complex_id	stop_lat	stop_lon
65	Clark St	bk061	40.697466	-73.993086
66	Borough Hall	bk068	40.693219	-73.989998
67	Hoyt St	bk097	40.690545	-73.985065
68	Nevins St	bk123	40.688246	-73.980492
69	Atlantic Av - Barclays Ctr	bk026	40.684359	-73.977666

	X	Y	id	name	address	zip	factype	facname	capacity	capname	bcode	xcoord	ycoord
0	1039079	234351	12000000001304	MARITIME COLLEGE AT FORT SCHUYLER (SUNY)	6 Pennyfield Ave	10465	1201	SUNY - The State University of New York	1799.0	Enrollment	36005	1039079.0	234351.0
1	999458	177948	12000000000085	HEALTH SCIENCE CENTER AT BROOKLYN (SUNY)	450 Clarkson Ave	11203	1201	SUNY - The State University of New York	1865.0	Enrollment	36047	999458.0	177948.0
2	985698	211585	12000000000022	FASHION INSTITUTE OF TECHNOLOGY (SUNY)	227 W 27 St	10001	1201	SUNY - The State University of New York	9764.0	Enrollment	36061	985698.0	211585.0
3	989188	214157	12000000000023	NYS COLLEGE OF OPTOMETRY (SUNY)	33 W 42 St	10036	1201	SUNY - The State University of New York	363.0	Enrollment	36061	989188.0	214157.0
4	1009244	251859	12000000001458	BRONX COMMUNITY COLLEGE (CUNY)	2155 University Ave	10453	1202	CUNY - The City University of New York	11506.0	Enrollment	36005	1009244.0	251859.0

	Tips_count	Price tier	Likes_count	Rating
ID
594277b586f4cc0f251fc389	96	2	731	9.3
59da9590e1f0aa52976b8f35	30	1	145	8.8
5a87210eda5ede7ae86f82c6	4	2	24	8.0
53d1a481498e1fac7c37f675	15	2	37	8.1
4b6b6ad0f964a52066072ce3	15	2	38	8.8
53b36289498e528b6cad6624	39	1	165	8.1
5946d138e2da1964625ca9b2	10	2	46	8.6
594416bca4ba7c683f5492df	3	1	36	8.5
50478b43e4b05a8c89d8198c	27	1	96	8.4
479ccb47f964a5206b4d1fe3	66	2	140	8.4
52b204c0498efe3c3be32d91	32	1	140	9.2
5c56083dc58ed7002c3255d0	3	2	10	8.7
5c3ebf15c0f163002ce6db6f	2	2	13	8.1
4ac00013f964a520689320e3	45	2	92	8.5
55664aac498e10d4e5769e5e	19	1	82	9.3
583c7fa58cfe547612b47b93	22	2	207	9.0
54b4026b498e73880a40a8d7	80	3	420	9.0
4e793558aeb79f7dab5c7f6f	114	2	416	9.0
5012043564a4944f5c47738a	56	2	172	9.0
4dc5a9c8e4cd169dc6532c66	121	3	306	9.2
4dcc0c3052b17cba4fb62352	38	1	114	7.8
530931a5498e4079544a5f13	112	3	550	9.0
3fd66200f964a520eae81ee3	31	2	87	8.4
550c4a17498e1d6612e9dffc	12	2	81	8.8
3fd66200f964a520ece81ee3	87	2	219	8.4
43eb685af964a520382f1fe3	95	2	333	9.0
51b0ceff2d4e2ee801fd4ab2	97	1	300	8.8
5942d5c94c9be62191b036fa	12	1	89	9.1
49d3773df964a520fb5b1fe3	254	3	691	9.1
4a635ce2f964a520dbc41fe3	48	1	100	8.4
59456df7a4ba7c683f19f0e7	9	1	31	8.1
4b5e4c9df964a520648829e3	310	2	737	8.8
4a748cd9f964a52092de1fe3	63	1	133	8.7
4cbf097828d176b0a7c6226e	233	3	732	9.2
49e63b62f964a52027641fe3	88	3	193	8.9
49f9dd95f964a5208d6d1fe3	60	2	195	8.7
49f3261bf964a520696a1fe3	92	2	211	8.1
4e1e23631838bed7eb6a8792	173	1	532	8.8
4f3265a119836c91c7d3c6f7	6	2	11	8.7